Machine LearningAdvanced
End-to-end implementation of a decoder-only transformer architecture with custom KV cache kernels and FlashAttention-2 integrations.
#PyTorch#Transformers#CUDA#LLM
38.4kRepo
Sub-second full-text search across curated tutorials, mini-projects, and machine learning models.
Showing 20 of 1,048,576 verified open-source items
Every Thursday, we deliver 10 exceptional open-source systems, engineering tutorials, and architecture teardowns directly to your inbox.